Assessing statistics attitudes among college students: Psychometric properties of the Italian version of the Survey of Attitudes toward Statistics (SATS)

doi:10.1016/j.lindif.2008.10.008

Learning and Individual Differences

Volume 19, Issue 2, June 2009, Pages 309-313

https://doi.org/10.1016/j.lindif.2008.10.008 Get rights and content

Abstract

The aim of the present research was to test the psychometric properties of the Italian version of the Survey of Attitudes Toward Statistics (SATS) in order to acquire a valid instrument to measure attitudes inside the Italian educational context. The SATS was administered to Italian-speaking undergraduate students enrolled in introductory statistics courses. A confirmatory factor analysis validated the four-factor structure of the scale, and good indices for both reliability and validity were obtained. A multi-group analysis confirmed the time invariance (beginning vs. end of the course) of the SATS, with a minor exception. Given the invariance across time, latent mean structures analysis revealed small to large significant changes in the four attitude dimensions. The present study provided evidence for the validity and reliability of the SATS when applied to the Italian learning framework.

Section snippets

Participants

The original sample consisted of 487 psychology students attending the University of Florence in Italy, who enrolled in an undergraduate introductory statistics course. Participants age ranged from 19 to 48 (Mean = 21.4, SD = 3.35, and Median = 20), most of the participants were women (84.1%). All students participated on a voluntary basis. The ATS scale (Wise, 1985) was administered to a sub-sample made up of 457 of which 313 students completed both the pre-course version of the SATS and ATS, and

Results

The SATS (both in the pre- and post-version) and the ATS responses to negatively worded items were reversed with higher ratings always representing more positive attitudes. Cases with missing values were eliminated (listwise method, n = 33 in the pre-SATS sample, n = 19 in the post-SATS sample). Outliers were identified using the box-plot procedure (Chambers, Cleveland, Kleiner, & Tukey, 1983): cases with scores lying outside the range defined upon the quartiles [Q₁ − 1.5 (Q₃ − Q₁); Q₁ − 1.5 (Q₃ − Q₁)]

Discussion

Aiming to investigate attitudes toward statistic in Italian-speaking students, an Italian version of the SATS was developed and its psychometric properties were tested. The structure of the Italian version of the SATS was established through a cross-validation procedure, which demonstrated that, as in the original English version, the four-factor structure provided a good fit for the observed interrelationships among parcels of items. All parcels loaded strongly and significantly on their

References (35)

FinneyS.J. et al.
Self-efficacy beliefs in college statistics courses
Contemporary Educational Psychology
(2003)
ArbuckleJ.L.
Amos 5.0 [computer software]
(2003)
BentlerP.M.
EQS: Structural equations program manual
(1995)
ByrneB.M.
Measuring adolescent self-concept: Factorial validity and equivalency of the SDQ III across gender
Multivariate Behavioural Research
(1988)
ByrneB.M.
Structural Equation Modeling with AMOS
(2001)
CarmonaJ.
Una revisión de las evidencias de fiabilidad y validez de los cuestionarios de actitudes y ansiedad hacia la estadística
Statistics Education Research Journal
(2004)
CarmonaJ. et al.
Mathematical background and attitudes toward statistics in a sample of Spanish college students
Psychological Reports
(2005)
CashinS.E. et al.
The Survey of Attitudes Toward Statistics Scale: A construct validity study
Educational and Psychological Measurement
(2005)
ChambersJ.M. et al.
Graphical Methods for Data Analysis
(1983)
ChiesiF. et al.
La scala Attitude Toward Statistics (ATS) di Wise: un primo contributo alla versione italiana
Bollettino di Psicologia Applicata
(2004)

CookL.L. et al.

An assessment of the dimensionality of three SAT-verbal test editions

Journal of Educational Statistics

(1988)

CudeckR. et al.

Cross-validation of covariance structures

Multivariate Behavioural Research

(1983)

DauphineeT.L. et al.

Survey of attitude Toward Statistics: Factor structure and factorial invariance for female and males

Structural Equation Modeling

(1997)

EcclesJ.S. et al.

In the mind of the actor: The structure of adolescents' achievement task value and expectancy-related beliefs

Personality and Social Psychology Bulletin

(1995)

FeldtL.B. et al.

Statistical tests and confidence intervals for Crombach's coefficient alpha (Iowa Testing Programs Occasional Papers No. 33), (ERIC Document Reproduction Service No. ED291755)

(1986)

GalI. et al.

Monitoring attitudes and beliefs in statistics education

GribbonsB.C. et al.

Levels of aggregations in higher level confirmatory factor analysis: Application for academic self-concept

Structural Equation Modeling

(1998)

Cited by (63)

Mean differences and individual changes in nursing students' attitudes toward statistics: The role of math background and personality traits
2021, Nurse Education in Practice
Nursing students have compulsory statistics courses in their degree program, but they usually have negative attitudes toward statistics that may hinder their learning. The present study aims: (i) to investigate nurse students’ attitudes toward statistics and the relationships with mathematical background and personality traits; (ii) to stress individual changes in attitude that occur during the course; and (iii) to explore if mathematical background and personality traits influence these changes. We adopted a one-group pre-post survey study. Fifty nursing students were enrolled in an introductory statistics course at the Sapienza University of Rome in Italy during the 2018–2019 academic year. Participants were surveyed at the beginning and end of the course administering a multidimensional measure of attitude toward statistics. Multiple regression analyses were run to establish the relative impact of mathematical background and Big Five personality factors on attitude components, as well as the changes in these attitude components. Results confirmed the predictive role of mathematic competence on some attitude dimensions and showed that also personality traits influenced attitudes toward statistics. However, the observed changes in attitudes during the course were minimally influenced by these factors. Findings suggest that course pedagogy can enhance students’ attitudes regardless their mathematical background and personality.
Examining developmental relationships between utility value, interest, and cognitive competence for college statistics students with differential self-perceived mathematics ability
2021, Learning and Individual Differences
Citation Excerpt :
Pre-course students' statistics attitudes comprising several motivation-related constructs were assessed using a Dutch translation of the pre-version of the Survey of Attitudes Toward Statistics-36 (SATS-36; Vanhoof, 2010). Evidence related to the sound psychometric properties of the SATS has accumulated (Chiesi & Primi, 2009; Nolan et al., 2012), particularly at the item level (e.g., Xu and Schau, 2019; Persson et al., 2019). All SATS items were rated on seven-point Likert scale (1 = Strongly Disagree, 4 = Neutral/No opinion, 7 = Strongly Agree).
Although there is a growth of interest in student motivation in higher education, not enough is known about the developmental relationships between motivational components among college students. In this study, a sample of 503 students in a five-year combined Bachelor's/Master's degree program at a Belgian university reported their statistics cognitive competence, utility value, and interest at the beginning of or prior to each of three progressively more advanced statistics courses. Results from latent growth curve analysis (LGCA) revealed that students' self-perceived mathematics ability at the beginning of the first course was positively associated with initial levels of statistics interest, utility value, and cognitive competence but not with their developmental trends. Moreover, results from multigroup LGCA suggested that students who reported lower mathematics ability began the first statistics courses with lower initial levels of interest, utility value and cognitive competence, but experienced development in these constructs at similar rates as their peers reporting higher mathematics ability. We also found that rates of growth in cognitive competence or value were positively associated with rates of decrease in interest. Implications of these findings for instructional practice are discussed.
Assessing the Statistical Anxiety Rating Scale as applied to prospective teachers in an Israeli Teacher-Training College
2020, Studies in Educational Evaluation
Citation Excerpt :
Evans (2007) and Hannigan et al. (2013) assume only three similar dimensions: (a) Statistics relevance; (b) Value, and (c) Difficulty and avoid the dimensions of affect and cognitive competence. However, the three most common classical dimensions of statistics anxiety as an attitude (Chiesi & Primi, 2009; Estrada et al., 2011; Onwuegbuzie, 2004) are (a) the Affective aspect, (b) the Cognitive aspect, both identified by Schau et al. (1995), and (c) the Behavioral aspect, referring to a person’s inclination to act in a particular way when studying statistics. This mapping is also common in the measurement of similar psychological phenomena, such as attitudes toward mathematics (Estrada et al., 2011) and the Beck Scale (Beck Depression Inventory II) for the measurement of depression (Cohen, 2008).
Statistics literacy is a growing part of our lives. However, around 75% of social science undergraduate students experience moderate to high levels of statistics anxiety, characterized by troubling thoughts, stress and an aversion to statistical content or to learning situations using statistics. The aims of the study were to examine the psychometric properties of the Hebrew version of the Statistical Anxiety Rating Scale, and to unfold the underlying structure of the construct as perceived by 163 pre-service teachers in an Israeli college of education. Multidimensional scaling analysis (MDS) confirms the confirmatory factor analysis (CFA) findings but highlights the importance of identifying the context of anxiety (statistics course vs. statistics as a distinct discipline) and the locus of anxiety (internal vs. external). The findings can help teacher educators better monitor their students’ statistics anxiety and assist them in implementing appropriate ways of reducing it.
Assessing attitudes towards biostatistics education among medical students: Adaptation and preliminary evaluation of the Chinese version survey of attitudes towards statistics (SATS-36)
2023, Research Square
Telling My Story: Applying Storytelling to Complex Economic Data
2023, Eastern Economic Journal
Examining statistical literacy, attitudes toward statistics, and statistics self-efficacy among applied linguistics research students in China
2023, International Journal of Applied Linguistics (United Kingdom)

View all citing articles on Scopus

View full text

Assessing statistics attitudes among college students: Psychometric properties of the Italian version of the Survey of Attitudes toward Statistics (SATS)

Abstract

Section snippets

Participants

Results

Discussion

Contemporary Educational Psychology

Amos 5.0 [computer software]

EQS: Structural equations program manual

Measuring adolescent self-concept: Factorial validity and equivalency of the SDQ III across gender

Multivariate Behavioural Research

Structural Equation Modeling with AMOS

Una revisión de las evidencias de fiabilidad y validez de los cuestionarios de actitudes y ansiedad hacia la estadística

Statistics Education Research Journal

Mathematical background and attitudes toward statistics in a sample of Spanish college students

Psychological Reports

The Survey of Attitudes Toward Statistics Scale: A construct validity study

Educational and Psychological Measurement

Graphical Methods for Data Analysis

La scala Attitude Toward Statistics (ATS) di Wise: un primo contributo alla versione italiana

Bollettino di Psicologia Applicata

An assessment of the dimensionality of three SAT-verbal test editions

Journal of Educational Statistics

Cross-validation of covariance structures

Multivariate Behavioural Research

Survey of attitude Toward Statistics: Factor structure and factorial invariance for female and males

Structural Equation Modeling

In the mind of the actor: The structure of adolescents' achievement task value and expectancy-related beliefs

Personality and Social Psychology Bulletin

Statistical tests and confidence intervals for Crombach's coefficient alpha (Iowa Testing Programs Occasional Papers No. 33), (ERIC Document Reproduction Service No. ED291755)

Monitoring attitudes and beliefs in statistics education

Levels of aggregations in higher level confirmatory factor analysis: Application for academic self-concept

Structural Equation Modeling